DE4.7.1 — Coding Images of Music MUSICNETWORK The Interactive-Music Network DE4.7.1 CIMS: Coding Images of Music Sheets Version: 2.2 Date: 03/02/04 Responsible: UNIVLEEDS Project Number: IST-2001-37168 Project Title: The Interactive-Music Network Deliverable Type: Doc./HTML Visible to the Working Groups: yes Visible to the Public: yes Deliverable Number: DE4.7.1 Contractual Date of Delivery: M18 Actual Date of Delivery: 03-02-2004 Title of Deliverable: CIMS: Coding Images of Music Sheets Work-Package contributing to the Deliverable: WP4 Nature of the Deliverable: Public Working Group: CIMS (Music Imaging WG) Author(s): Kia Ng, Jerome Barthelemy, Bee Ong, Ivan Bruno, Paolo Nesi Abstract: This document reports the applications and practices in the domain of coding images of music sheets (music imaging), which include music sheet digitisation, recognition, restoration, and others. It reports hardware and software related to music imaging, with discussions on main obstacles and approaches to evaluate state of the art OMR system. Keyword List: Music imaging, music digitisation, sheet music, image processing, scanner, optical music recognition, OMR, optical music restoration, multimedia, image
Transcript
DE4.7.1 — Coding Images of Music
MUSICNETWORK
The Interactive-Music Network
DE4.7.1 CIMS: Coding Images of Music Sheets
Version: 2.2 Date: 03/02/04 Responsible: UNIVLEEDS Project Number: IST-2001-37168 Project Title: The Interactive-Music Network Deliverable Type: Doc./HTML Visible to the Working Groups: yes Visible to the Public: yes Deliverable Number: DE4.7.1 Contractual Date of Delivery: M18 Actual Date of Delivery: 03-02-2004 Title of Deliverable: CIMS: Coding Images of Music Sheets Work-Package contributing to the Deliverable: WP4 Nature of the Deliverable: Public Working Group: CIMS (Music Imaging WG) Author(s): Kia Ng, Jerome Barthelemy, Bee Ong, Ivan Bruno, Paolo Nesi Abstract: This document reports the applications and practices in the domain of coding images of music sheets (music imaging), which include music sheet digitisation, recognition, restoration, and others. It reports hardware and software related to music imaging, with discussions on main obstacles and approaches to evaluate state of the art OMR system. Keyword List: Music imaging, music digitisation, sheet music, image processing, scanner, optical music recognition, OMR, optical music restoration, multimedia, image
DE4.7.1 — Coding Images of Music
MUSICNETWORK Project 2
Table of Content
1 EXECUTIVE SUMMARY AND REPORT SCOPE .............................................................................................. 3
6 OMR.............................................................................................................................................................................. 6 6.1 COMMERCIAL OMR SYSTEMS .............................................................................................................................. 6
7 OMR EVALUATION................................................................................................................................................. 7 7.1 OBSTACLES ........................................................................................................................................................... 7 7.2 THE OMR QUICK-TEST ........................................................................................................................................ 8 7.3 EVALUATION OF PERFORMANCE BASED ON COMPLETE MUSIC SYMBOLS AND RELATIONSHIPS RECONSTRUCTION 9
8 MUSIC IMAGE RESTORATION.......................................................................................................................... 13
9 APPLICATIONS AND FUTURE DIRECTIONS ................................................................................................ 14
1 Executive Summary and Report Scope This document reports the applications and practices in the domain of coding images of music sheets (music imaging), which include music sheet digitisation, optical music recognition (OMR) and optical music restoration. With a general background of Optical Music Recognition (OMR), the report discusses typical obstacles in this domain and reports currently available commercial OMR software. It reports hardware and software related to music imaging and discusses steps required to evaluate the state of the art OMR system. Besides the main focus on the transformation from images of music scores to symbolic format (for printed and handwritten music notation), this document also reports music image restoration and the application of music imaging techniques for graphical preservation and potential applications for cross-media integration. 2 Introduction The document explore issues on the digitisation, restoration and automatic transcription of music documents; converting paper-based music document into machine readable formats, in order to explore effective use of the latest interactive and multimedia technologies for cultural heritage restoration and preservation of musical documents, such as printed music scores, handwritten manuscripts and ancient music scores. With the advancements of digitisation and information technologies, document analysis and optical character recognition technologies are now widely used, from form processing to handwritten address recognitions. As we know, document imaging, analysis and understanding is extremely complex, not to mention the additional complexities inherent to Music notation. There are a vast amount of invaluable paper-based heritage, including printed music scores and handwritten manuscripts, that are deteriorating over time due to natural decaying of paper and chemical reaction (e.g. printing ink and paper), similar to many other paper-based items in library and museum archives. In order to introduce interactive multimedia music capabilities and functionalities, machine readable representation is required, and hence one of the main steps is to create digital version of these paper-based heritage materials for further processing (restoration, encoding, recognition etc) in order to allow long term preservation and wider and more effective distributions. Various efforts have been focused on this issue in order to preserve the record of our heritage. For example, manual and highly skill paper-splitting technique used to conserve Bach’s manuscripts [Porck & Teygeler, 2000; Wächter et al., 1996]. 3 Background Digitisation has been commonly used as a possible tool for preservation. Although the digital copy may not conserve the original document, it can preserve the data in the document, with the advantage of easy duplications, distribution and digital processing. Optical Music Recognition (OMR), also commonly known as OCR for Music (Optical Character Recognition for Music) was first attempted in the 60s, and since then there have been a wide range of research and developments in this interdisciplinary domain. Currently there are various commercially available products as well as research systems for OMR. OMR system transforms paper-based printed music scores and handwritten music manuscripts, into a machine-readable symbolic format, and an optical music restoration system to reconstruct small discontinuities and imperfection in the musical writings, including broken stems and stave lines. An idealise system which could reliably “read” and “understand” music notations could provide a wide range of applications for interactive multimedia music, bringing paper-based music to the new multimedia era. OMR was first attempted over thirty years ago [Pruslin, 1966]. It has received much attention over the last fifteen years [Bainbridge & Wijaya, 1999; Bellini et al., 2001; Bruno & Nesi 2002; Ng & Boyle, 1992; Ng, 1995; Ng et al., 1999; Ng, 2002; etc, see Section “OMR Bibliography”], and there are currently a number of commercially available packages, such as capella-scan [capella-scan], Optical Music easy Reader [OMeR], PhotoScore [PhotoScore], SharpEye [SharpEye], SmartScore [SmartScore] and Vivaldi Scan [Vivaldi Scan].
DE4.7.1 — Coding Images of Music
MUSICNETWORK Project 4
However there are still much room for improvements in many aspects. Reviews and background on the development of various OMR systems can be found in Bainbridge & Carter [1997], Blostein & Baird [1992] and Selfridge-Field [1994]. An online bibliography on OMR can be found at the Interactive MUSICNETWORK website (http://www.interactiveMUSICNETWORK.org) and http://www.kcng.org/omrbib/ 4 Obstacles, Music Notation Optical Character Recognition (OCR) is perhaps the best known related document image processing problem, but OMR can be critically different. The visual problem might seem simple since writing is normally black on white paper. However, OMR introduces an additional layer of complexity due to the wide range of possible shape variation resulted from inter-connections and groupings of symbols. Furthermore there may be other symbols (e.g. expressive signs, fingerings, bowing, texts, etc.) that are positioned around and sometime overlaid part other music symbols. For example, a tie crossing a stem or touching a note-head. Music Notation is inherently opened ended. Even if generally considered as stable for the period of XVIIIth and XIXth centuries in the Western world, there are several exceptions, such as “unmeasured notation” (for cadenzas and so on), approximate rhythmic notation (several examples can be found in works of authors like Chopin, Schumann or Mendelssohn), or slight enhancements to traditional notation (slurs without ending note, non canonical time signatures…). In the earlier centuries, with neumatic or Gregorian notation, music notation was very far of a standardized system, and in the XXth century, music notation has exploded, and is noticeably far from that model commonly known as Common Western Music Notation. Direct recognition of musical symbols is difficult due to the design of the notation. In general, OMR system uses divide-and-conquer approaches to separate musical features before recognition. For example, stave lines are detected and marked before/after note-head in order to separate one feature from the other. Basic musical syntax (e.g. time-signature) and domain-knowledge enhancement such as rhythmical analysis have been explored to improve recognition performance. Fahmy & Blostein [1998, 1994] propose a graph-rewriting approach for OMR enhancement. Stückelberg et al. [1997] propose an architecture for OMR with high-level domain knowledge and Stückelberg & Doermann [1999] explore probabilistic reasoning for musical score recognition. Coüasnon [2002] comments that existing OMR software is not suitable for industrial context due to time consuming and tedious manual proof reading, and proposes a system that is capable of self-diagnostic to detect error [Coüasnon and Rétif, 1995]. The paper discusses the application of musical knowledge of music writing to enhance OMR processing and recognition using DMOS (Description of MOdification of Segmentation), a generic recognition approach for structured document analysis with grammatical formalism EPF (Enhanced Position Formalism). 5 Music Digitisation
5.1 Hardware Nowadays, document digitisation systems such as optical flatbed scanners are widely available. There are a wide range of commercial products from manufacturers such as Fujitsu, Agfa, HP, Cannon, Epson, UMAX, Microtek, Visioneer and many more. Currently available commercial products are equipped with USB, parallel or SCSI interfaces. Some of these products support dual-interfaces. Many of these products are capable of more than 600 d.p.i. (dot per inch) optical scan resolution with grey or up to 48-bit colour depth which surplus general requirement for OMR processing. Increasingly digital photo-copiers are also equipped with optical scanner which provides high-speed digitisation. Examples include products from Ricoh and Canon.
DE4.7.1 — Coding Images of Music
MUSICNETWORK Project 5
Drum scanners are less commonly being used in this domain. Besides professional flatbed scanners (such as Creo Scitex, Heidelberg and others), industrial music imaging applications for archiving (bitmap images) also use a digital-camera-back or digital-camera with a copy-stand setup which range from a simply board for document placement to include fully pneumatically controlled book cradle system as well as complex robotic control automatic page-turning system. Examples of overhead-scanning products include:
Company Product Notes URL Kirtas Technologies, Inc. (USA) APT BookScan 1200 World's first automatic book
scanner http://www.kirtas-tech.com
4DigitalBooks "DIGITIZING LINE" Automatic digitizing system http://www.4digitalboo
ks.com
Zeutschel GmbH various MONISCAN models
Large format colour scanner OMNISCAN9000
http://www.zeutschel.de
Solar Imaging Systems, UK M3 & M5 digital camera systems
Maximum optical resolution 8192x12000 pixels
http://www.solar-imaging.com
Icam Archive Systems, UK GUARDIAN Various models including Guardian which uses Phase One camera backs
http://www.icamarchive.co.uk
Konica Minolta Minolta PS7000 book scanner up to A2, 256 greyscales http://www.minoltaeuro
pe.com/...
InfoSys GmbH alpha librae up to 900 pp/hour, greyscale & colour model
http://www.infosys-scanner.de/indexE.html
ImageWare Components GmbH Bookeye products Oversize formats up to 350 x 720 x 470 mm
http://www.bookeye.com
Imaging Business Solutions SMA ScanFox A1 and A2 http://www.imagingbusiness.co.uk
5.2 Digitisation There are many digitisation related projects and institutions which have produced good set of guidelines or publications related to these issues. These include:
• MINERVA, http://www.minervaeurope.org • PULMAN, http://www.pulmanweb.org • AHDS (Arts and Humanities Data Service), UK, http://www.ahds.ac.uk • British Library, http://www.bl.uk/services/preservation/freeandpaid.html • CLIR (Council on Libraries and Information Resources), Building and sustaining digital collections:
models for libraries and archives, http://www.clir.org • DLF (Digital Library Federation), Digital library standards and practices,
http://www.diglib.org/standards.htm • Library of Congress, A Framework of Guidance for Building Good Digital Collections ,
http://www.nap.edu/catalog/9940.html • UNESCO/ICA/IFLA, Guidelines for digitization projects for collection and holdings in the public
domain, particularly those held by libraries and archives http://www.ifla.org/VII/s19/pubs/digit-guide.pdf
• DI.MU.SE project (Ministero per i Beni e le Attività Culturali and Palatina Library of Parma) provided guidelines for the digitalisation of 150.000 music manuscript pages. http://www.bibpal.unipr.it
• and others As with other document imaging processes such as Optical Character Recognition (OCR), OMR is not particularly demanding on currently available optical document scanners. Typically, for sheet music, 300 d.p.i optical resolution and 8-bit grey is sufficient for the purpose of OMR [Selfridge-Field 1994]. Fujinaga & Riley (2002) reported that 600dpi is a sufficient resolution for all significant details. The paper suggested that further increase in resolution is not necessary for OMR. Generally, the first process in a document analysis system is to threshold a given grey input image into a binary image. Some systems used binary input images produced by the digitiser. Regarding the Italian DI.MU.SE project, the digitalisation parameters for the music manuscript was fixed for the master copy in 300 d.p.i optical resolution with colour depth at 24 bit RGB, TIFF format. 6 OMR
6.1 Commercial OMR Systems Current, there is a number of commercially available OMR software. No comprehensive comparative study has been carried out, and hence this is urgently required. In order to provide unambiguous comparative study between different software, terminology for all musical primitives (e.g. note-head, stem, etc) has to be standardise. For a non-bias survey of OMR software, a representative and sufficiently large ground-truth dataset of music sheets containing different style of fonts, density, sizes and page layout has to be collected. Simple recognition rate (as in Optical Character Recognition) does not offer good/meaningful measure for OMR system due mainly to the complex musical notation which may change (visually) depending on the contexts, and hence a good assessment matrix is require to provide meaningful assessment for OMR system. Current proposal include a 3 level approach with measurement at primitive-level, note-level and interpretation- (score-) level. Commercially available OMR systems include:
• BraeburnSoftware, Music Publisher system • Vivaldi Scan (derived from SharpEye) • Musitek, SmartScore • Scorscan of NPCImaging http://www.npcimaging.com/scscinfo/scscinfo.html • MIDI-Connections Scan, http://www.midi-connections.com/Product_Scan.htm
7 OMR Evaluation
7.1 Obstacles The Optical Music Recognition task is more complex than OCR. Despite to the availability of several commercial OMRs: SharpEye2, SmartScore, Photoscore, CapellaScan, etc., none of these is satisfactory in terms of precision and reliability. The efficiency declared by the each distributor is close to 90%, but this value is obtained only when quite regular music sheets are processed and the estimation is not always objective. In the character or face recognition field, there are many ground truth databases that enable recognition results to be evaluated automatically and objectively. At the present time, there is neither a standard database for music score recognition or a standard terminology. If a new recognition algorithm or system were proposed, it could not be compared with the other algorithms or systems since the results would have to be traditionally evaluated with different scores and different methods. Taking these facts into consideration, it is indispensable to make a master music score database that can be used to objectively and automatically evaluate the music score recognition system. At the same time a set of rules and metrics are needed in order to define what aspects have to be considered in the evaluation. In general, the currently available commercial OMR systems are linked to a music notational software. For example, PhotoScore outputs directly into Sibelius. It is not easy to access the performance of the OMR system alone without interaction with the interface provided by the notational software. That is to say, it is not always possible to output from the OMR system itself. This problem is even complicated by the lack of a commonly accepted standard for musical notation. All the currently available OMR systems offer capture facilities to communicate directly with the scanner. This is another complication since it is not easy to make sure that the inputted images for all OMR systems are exactly the same (pixel perfect). For the OMR assessment, file input is clearly preferred, however the complications here include:
• different input format support • different optimum resolution • different image depth requirement (e.g. 2-bit, 8-bit etc.) and different pre-processing approaches.
Besides differences of input and format representation, differences in output formats, due to the lack of a commonly accepted standard for musical notation, present another layer of complication. Generally output can only be obtained in the proprietary format of the music notation software – that is, the “.mus” or the Enigma format for the Finale software, the Sibelius format for the Sibelius software, and so on. Some of these formats, such as the “.mus” format, are undocumented, and some partially documented format such as the Enigma format are perpetually evolving and suffering of lacks of documentation. The NIFF format, which was designed at the very origin for the purpose of exchanging music notation between different music notation software, noticeably OMR and music notation software, is now used by very few notation software (see the MUSICNETWORK deliverable DE4.1.1, “music notation coding”), and some OMR software are not able to export in this format. In general, all music notation software could export in the MIDI format, but this format doesn’t capture all features of music notation, thus that format can be used only in a first approach. Noticeably, the MIDI format doesn’t capture rhythmic features, and output in MIDI from different music notation software could be slightly different depending on the music software. It would however be possible to set up a methodology based on the MIDI format, with a first step of export in MIDI, followed by an import in a reference software. It is not easy to compare results outputted in different formats due to their individual designs and capabilities. Due to these complications, “scan once, use many” methodology may not be easily applied.
DE4.7.1 — Coding Images of Music
MUSICNETWORK Project 8
Moreover, complexity of music notation is a supplementary challenge: the fundamental unit of music notation (the note) is itself a complex object, made of a note head, a stem, a flag or a beam, possibly an accidental and dynamic markings such as staccato dot. This complex object can be modified in its fundamental meaning - pitch and duration - by its environment (clef, key signature, time signature…). The results of the comparison can be distorted by errors in context which cannot be taken in account by the comparison test himself (tempo marking). Thus, a good methodology for comparing results must involve definition of different ratios for each kind of error, and is to be taken into account in the methodology definition. An error of clef, for example, would produce an error for each note in the MIDI output, while being easily corrected by just one correction in the notation format. These errors must be corrected at the earliest step possible, since errors done could induce bad corrections and artefacts introduced by software at a later step (for example, error in time signature could introduce completion of measures in notation software, by introduction of incorrect rests for completion of measures). In the same manner, a normalisation must be done at an earlier stage to correct possible errors of non-significant features such as tempo markings which could introduce differences in the final output. A proposed methodology for comparing OMR software would then involve the following steps:
1. Input of scan with different resolutions, different format support, different image depth 2. First step of correction for context errors: clef, time signature, key signature. 3. Normalization of context for non tested features: tempo marking 4. Output in music notation software 5. Second step of correction for context errors: clef, time signature, key signature (if not possible at an
earlier stage) 6. Normalization of context for non tested features: tempo marking (if not possible at an earlier stage) 7. MIDI export 8. MIDI import in a reference software – Normalization of context (tempo markings – dynamics – MIDI
instruments and parts) – first evaluation of rates 9. Correction of errors on the first-rated result, and generation (MIDI export) of a reference file 10. MIDI export
The comparison tests must be made:
By manual, human detection of errors at step 8 (MIDI import) described above. By automatic, software-based comparison of the results obtained at step 10 with the reference file
obtained in 9.
This methodology can only be applied to those basic features which are part of the MIDI standard, and cannot be applied to features which are not part of the standard, such as dynamic markings (hairpins, staccato, tenuto…).
7.2 The OMR Quick-Test The OMR “Quick-Test” (version 0.1) contains three pages of basic musical features including:
• time signatures • notes • beams • key signatures • clefs • note heads • accidentals • articulation • text • bar lines • flat beams • sloping beams • stave types • dynamics
DE4.7.1 — Coding Images of Music
MUSICNETWORK Project 9
• hairpins • rests • slurs and ties • triplet and tuplets • octava lines • pedal symbols • pedal lines
As mentioned above, most OMR systems use notational software front-ends. Notational software tends to use domain knowledge to interpret the data e.g. the number of beats in a bar given a known time-signature. To try and eliminate as far as possible the difficulties that misreading of this kind produces, the “Quick-Test” is designed with “correct” time to occupy each bar. The data gathered from the “Quick-Test” should result in a list of recognition problems allowing us to target features which consistently prove problematical for the software. This will then lead to a priority list of these features. Solving these problems will lead to more accurate initial interpretation of symbols and features, and therefore to less user-intervention. We also look forward to input from the content providers that have supplied us with images, as well as from the OMR system developers. Their comments will help in identifying areas that are important to address if OMR software is to become more commercially viable. Further details and information are available online at the Imaging WG section at the project website: http://www.interactiveMUSICNETWORK.org The proposed Quick-Test dataset is available for download at the Imaging WG section.
7.3 Evaluation of performance based on complete music symbols and relationships reconstruction
Beside the Quick-Test, the WG is also working on an approach for the assessment of OMR system intending to take into account the “importantness” of the each musica features and planning to offer a representative measure to measure the performance of OMR system. For further details, please see “Assessing Optical Music Recognition Tools” by I. Bruno, P. Bellini and P. Nesi available online at http://www.interactiveMUSICNETWORK.org/wg_imaging/upload/assessingopticalmusicrecognition_v1.0.doc. Three applications have been selected in order to compare the performance in the score recognition: SharpEye2 (Visiv), SmartScore (MusiTek) and O3MR (developed at the DSI – University of Florence). The set of complete symbols and relationships are listed and described in Fig. 1. This evaluation set is not exhaustive for all genre of music score, it could be extended in order to include more aspects (structural, symbolic, etc…). The proposed list is able to describe the monophonic music score and relationships, and the most important and frequent symbols. The relevance of each category is represented by a value of weight. The weights have been collected by interviewing a group of 13 people at the second MUSICNETWORK workshop. Definition of test set – The missing of a ground truth databases conditioned the choice of tests, to cope with this lack, seven images have been selected from the archive of collected images at the DSI. The test-cases can be found online at http://www.interactiveMUSICNETWORK.org/documenti/view_document.php?file_id=475. The chosen music scores have the following features: • Monophonic music. • Font variability. • Music symbols frequently used in the classic music repertory. • Variable density of music symbols. • Irregular groups (triplets, etc.). • Small note with or without accidentals (grace notes). • Different barlines (start and end refrain, end score, single barline and double barline).
DE4.7.1 — Coding Images of Music
MUSICNETWORK Project 10
• Clef and time signature change. • Ornaments (mordent, turn, and trill). • Slurs: single and nested. Result analysis - The Fig. 2 shows tables with global evaluations on the test set respectively of SmartScore, O3MR and SharpEye2, where: • The Total column reports the number of collected occurrences for each category. • The True column reports the percentage rate for correct symbols • The Add column reports the percentage rate for added symbols • The Fault column reports the percentage rate for incorrect symbols • The Miss column reports the percentage rate for missed symbols Tables show that: • SmartScore introduces errors in notes reconstruction and adds notes. It detects tuplets, but the main
tendency is to make mistakes. It has difficulty with slurs, time signature change and key signatures. • SharpEye 2 does not introduce notes, it has some problems with tuplets. In the grace notes detection, it
does not discriminate appoggiatura from acciaccatura, it considers only grace notes as appoggiatura. • The main limits for O3MR are due to the recognition of slurs, tuplets, grace notes and ornaments
symbols. It introduces wrong slurs due to a incorrect decomposition of symbols, whereas it adds less symbols than SmartScore. It obtained the best score in Time Signature, Key Signature and Clef recognition.
The recognition of notes and rests is the most important requirement that an OMR system has to respect. They are considered the most important music symbols, and their recognition has to be robust and to provide a high performance. Tables show the O3MR capability in recognising notes and rests. In particular, the recognition rate for rests is the highest, with a difference of 13.54% with SharpEye2 and 56.77% with SmartScore. Rests, added by O3MR (4.69%), are due to segmentation errors. This is a limit for the actual version of the O3MR system. The graphics reported in Fig.3 shows a global evaluation associated with each example. They represent respectively:
(i) The Weighted Percentage Reconstruction Rate: it takes into account weights associated with each music symbol and relationship.
Categories Weight Aim Note with pitch and duration
10 Evaluate the note reconstruction correctness in terms of pitch and duration.
Rests 10 Evaluate the recognition of rests. Note with accidentals
7 Evaluate the association of accidentals (sharp, flat, double sharp, natural, double flat) with a note.
Groups of beamed notes 10 Evaluate the capability in reconstructing beamed notes Time signature and time change 10 Evaluate the capability in identifying and reconstructing the time
indication by recognised numbers involved in the fraction. Key signature and key signature change
10 Evaluate the capability in identifying and reconstructing the key signature (tonality). The tonality is linked to the number of accidentals used in representing the key signature.
Symbols below or above notes
5 Evaluate the capability in identifying and linking ornaments symbols and accents (staccato, accent, turn, mordent, trill, tenuto, etc…).
Grace notes
5 Evaluate the capability in recognising grace notes: acciaccatura and appoggiatura are related to a single symbol while multiple grace notes define a group of notes. The multiple notes are considered a unique symbol.
Slurs and bends 7 Evaluate the reconstruction of horizontal symbols: slurs (and ties) and bends.
Augmentation dots 10 Evaluate the augmentation dots linking to notes. Clefs 10 Evaluate the recognition of clefs and clef changes. Irregular notes groups 10 Evaluate the capability in recognising tuplets. Number of measures
10 Evaluate the capability in recognising the bar line and the number of measures.
Number of staves 10 Evaluate the capability in recognising staves. Fig. 1 - List of complete symbols and relationship considered in the performance evaluation.
DE4.7.1 — Coding Images of Music
MUSICNETWORK Project 11
(ii) The Percentage Reconstruction Rate: in this case music symbols and relationships have the same relevance.
(iii) The Percentage Reconstruction Error: it considers missed, added and fault symbols. For this reason it represents a measure of the work has to be done to correct the reconstructed score.
This evaluation shows that SharpEye provides in general the best performance, whereas the O3MR is comparable with Smartscore. The O3MR obtained the best score with the example 5 (93.35%).
SmartScore SharpEye2 O3MR Complete Music symbols & Relationships
8 Music Image Restoration Besides recognition and translation into machine-readable symbolic representation, graphical representation of music scores and manuscripts can also be useful for various applications, including digital preservation and cross-media integration. The idea is to digitise, extract and encode the music graphically to preserve the look and feel of the original image from the paper-based input. This is particularly important for handwritten music manuscripts, since this approach preserves the writing style and enables scalable reconstruction and visualisation. Suitable vector graphics formats include:
• SVG (Scalable Vector Graphics); which is an XML-based 2D vector graphics file format, • MPEG 4 BIFS, • Postscript • Adobe PDF • Flash • and others
SVG (for Scalable Vector Graphics) is a standard (a recommendation) of the World Wide Web Consortium. SVG is a language for describing two-dimensional graphics and graphical applications in XML. Postscript is a language for description of a printed page. Developed by Adobe in 1985, it has become an industry standard for printing and imaging. The PDF (Portable Document Format) is based on Postscript, and on the ability of almost all software on major operating systems such as Windows or MacOS to generate postscript using their widely available Postscript printing device driver. The Flash format, developed by Macromedia, is mainly based on a vector graphics format, similar in functionalities to the Freehand format of the same vendor. It is a proprietary format, even if the specifications are available. MPEG BIFS (Binary Format for Scenes Description) makes possible to define so-called ”scenes“ consisting of several audiovisual objects which can be part of complex interactive multimedia scenarios. The individual objects are encoded and transmitted separately in a scene which is then composed after decoding of individual objects. Objects can be simple shapes such as circles, rectangles, text, or media such as AudioClip or MovieTexture, or even scripts. SVG and BIFS can be considered as something equivalent: SVG is XML-based, while BIFS, even if not XML-based, owns an equivalent in the XMT format which is the XML translation of BIFS. Flash is a proprietary format, even if publicly available (with some restrictions), and is subject to evolve without notice from the owner (Macromedia). Postscript, even if being page-based, can be a good choice since translation tools are available from Postscript to SVG (Adobe Illustrator), thus to BIFS by the mean of XMT. But Postscript is not absolutely free, while the SVG standard is absolutely free of patents or royalties, like every W3C standard. The SVG standard seems to be the best choice, for the following reasons;
- It can generate BIFS (and even Postscript, PDF or even Flash content). - SVG is an open standard, free of patents and royalties - Being developed by Adobe, the format is at the level of the state of the art.
For more information about vector graphics formats, please refer to the MUSICNETWORK deliverable 4.3.1, “multimedia standards for music coding”. Typical enhancements and restorations process include reconstructing broken stave lines and stems, and removing ink spillage and noise (see Figure below). Working at this level allows minor alteration such as this. However, this is not an effective approach for modifications involving larger interconnected features or alteration affecting multiple staffs. The advantage of optical music restoration is that the processes do not jeopardise the original layout of the scores, which have been optimised by the engravers, and normally represents the ideal visual configurations. Since the original spacing of the music is untouched, there is no large modification and hence it does not
DE4.7.1 — Coding Images of Music
MUSICNETWORK Project 14
require extensive proof reading. However, the process is only concerned with small and local modifications. Larger adjustments, for example insertions or deletions of a group of symbols cannot be fully automated without altering the original layout. No full recognition is necessary for this process and hence it does not provide multimedia functionalities such as playback or search. This process is robust and it can improve the visual qualities of the scores and manuscript for reprinting and archiving.
Example inputs After processing
9 Applications and Future Directions With an effective and robust OMR system, it can provide an automated and time-saving input method to transform paper-based music scores into a machine readable representation, for a wide range of music software, in the same way as OCR is useful for text processing applications. Besides direct applications, such as playback, musical analysis, re-printing, editing and digital archiving, OMR would enable efficient translations, for example, to Braille notations [Dancing dots] or other non-western musical notations. It could provide better access and widen participation of music and at the same time introduce new functionalities and capabilities with interactive multimedia technologies and provide digital preservation of this invaluable paper-based cultural heritage. With graphical reconstruction processes, paper-based music sheets can be digitised with the original visualisation with the capabilities of cross-media integration , extending useful functionalities for usages and applications in edutainment, long term preservation and archiving as well as widening accessibilities and participations. Typical applications in the field of cross-media (multimedia) integration includes the following:
- Association of scores and audio performance, with automatic synchronization. - Association of scores and video excerpts (showing details on execution) - Association of scores and other visualisations, such as musical summaries, sonagrams… - Hyperlinking (adding links to graphic symbols in scores) - Convergence with audio technologies (remix, spatialisation…) - Content-based queries, and web-based access to music (query by humming, query by
example…)
DE4.7.1 — Coding Images of Music
MUSICNETWORK Project 15
- Use of the score as a reference point for studies on expressive rendering (comparison of renderings from different performers), and use of score for expressive rendering using audio synthesis software
Association of scores and musical performance can actually be made manually, but in the case of a vector-graphics based score, an automatic process can be envisaged, in the near future for monophonic audio, and in a mid-term future for polyphonic music, with the progress of automatic voice separation. Based on this All these applications have direct application in the field of education as well as in the field of music practice. For more details on cross-media integration, see the MUSICNETWORK deliverable 4.3.1, “multimedia standards for music coding” 10 References Bainbridge, D., & Bell, T. (2001). The Challenge of Optical Music Recognition. Computers and the
Humanities, 35, 95-121. Bainbridge, D., & Carter, N. (1997). Automatic Recognition of Music Notation. Bunke, H. and Wang, P.
(Eds.), Handbook of Optical Character Recognition and Document Image Analysis, World Scientific, 557–603.
Bainbridge, D., & Wijaya, K. (1999). Bulk Processing of Optically Scanned Music. Proceedings of the 7th International Conference on Image Processing and Its Applications, 474–478.
Bellini, P., Bruno, I., & Nesi, P. (2001). Optical Music Sheet Segmentation. Proceedings of the First International Conference on WEB Delivering of MUSIC, 183–190.
Bellini, P., & Nesi, P. (2001). Wedelmusic Format: An XML Music Notation Format For Emerging Applications. Proceedings of the First International Conference on WEB Delivering of MUSIC, 79–86.
Blostein, D., & Baird, H. S. (1992). A Critical Survey of Music Image Analysis. Baird, H.S., Bunke, H. and Yamamoto, K. (Eds.), Structured Document Image Analysis, Springer-Verlag, Berlin, 405–434.
Bruno, I., & Nesi, P. (2002). Multimedia Music Imaging: Digitisation, Restoration, Recognition and Preservation of Music Scores and Music Manuscripts, 1st MUSICNETWORK Open Workshop, Darmstadt, Germany, 2002.
capella-scan (n.d.). capella Software, Retrieved February 9, 2003, from http://www.whc.de/ Carter, N.P. (1992). Segmentation and Preliminary Recognition of Madrigals Notated in White Mensural
Notation. Machine Vision and Applications, 5(3): 223-30. Carter, N.P. (1994). Conversion of the Haydn Symphonies into Electronic Form using Automatic Score
Recognition: A Pilot Study, in L.M. Vincent & T. Pavlidis (eds.), Proceedings of the SPIE – Document Recognition, 2181, 279–290.
Carter, N.P., & Bacon, R.A. (1990). Automatic Recognition of Music Notation. Proceedings of the International Association for Pattern Recognition Workshop on Syntactic and Structural Pattern Recognition, 482.
Choudhury, G.S., DiLauro, T., Droettboom, M., Fujinaga, I., Harrington, B. & MacMillan, K. (2000). Optical Music Recognition System Within a Large-Scale Digitization Project. International Conference on Music Information Retrieval.
Cooper, D., Ng, K.C., & Boyle, R.D. (1997). An extension of the MIDI file format: expressive MIDI – expMIDI. Selfridge-Field, E. (Ed.), Beyond MIDI: The Handbook of Musical Codes, MIT press, 80–98.
Coüasnon, B. and Rétif, B. (1995), Using a Grammar for a Reliable Full Score Recognition System, Proceedings of the International Computer Music Conference (ICMC), 187–194.
Coüasnon, B. (2002), Improving Optical Music Recognition, Position paper, First MUSICNETWORK Open Workshop, Darmstadt, Germany, 2002.
Dancing dots (n.d.). Goodfeel Braille Music Translator, Retrieved August 8, 2002, from http://www.dancingdots.com
Fahmy, H., & Blostein, D. (1994). A Graph-Rewriting Approach to Discrete Relaxation: Application to Music Recognition. Proceedings of the SPIE, 2181, 291-302.
Fahmy, H., & Blostein, D. (1998). A Graph-Rewriting Paradigm for Discrete Relaxation: Application to Sheet-Music Recognition. International Journal of Pattern Recognition and Artificial Intelligence, 12(6), 763-99.
DE4.7.1 — Coding Images of Music
MUSICNETWORK Project 16
Fujinaga, I. (1988). Optical Music Recognition using Projections. Master Thesis, McGill University, Montreal, Canada.
Fujinaga, I., Alphonce, B. & Pennycook, B. (1992). Interactive Optical Music Recognition. Proceedings of the International Computer Music Conference, 117-20.
Fujinaga, I. (1996). Exemplar-based learning in adaptive optical music recognition system. Proceedings of the International Computer Music Conference, 55-6.
Fujinaga, I. (2001). An Adaptive Optical Music Recognition System. David Greer (ed.). Musicology and Sister Disciplines. Past, Present, Future: Proceedings of the 16th International Congress of the International Musicological Society. Oxford, Oxford University Press.
Fujinaga, I., & Riley, J. (2002). Digital Image Capture of Musical Scores. Proceedings of the 3rd International Conference on Music Information Retrieval (ISMIR 2002), IRCAM – Centre Pompidou, Paris, France.
Gezerlis, V.G., & Theodoridis, S. (2000). An Optical Music Recognition System for the Notation of the Orthodox Hellenic Byzantine Music, International Conference of Pattern Recognition (ICPR-2000), Barcelona, Spain.
Good, M. (2002). MusicXML in Practice: Issues in Translation and Analysis. Proceedings of the 1st International Conference MAX 2002: Musical Application Using XML, Milan, 47–54.
Good, M. (2001). MusicXML for Notation and Analysis. In Hewlett, W.B. and Selfridge-Field, E. (Eds.), The Virtual Score: Representation, Retrieval, Restoration, MIT Press, Cambridge, MA, Computing in Musicology 12, 113–124.
Hoos, H.H., Hamel, K.A., Renz, K., & Kilian, J. (1998). The GUIDO Music Notation Format - A Novel Approach for Adequately Representing Score-level Music. Proceedings of the International Computer Music Conference, 451–454.
Matsushima, T., Harada, T., Sonomoto, I., Kanamori, K., Uesugi, A., Nimura, Y., Hashimoto, S., & Ohteru, S. (1985). Automated recognition system for musical score: The vision system of WABOT-2. Bulletin of Science and Engineering Research Laboratory, Waseda University.
Miyao, H., & Haralick, R. M. (2000). Format of ground truth data used in the evaluation of the results of an optical music recognition system. IAPR Workshop on Document Analysis Systems.
Ng, K.C., & Boyle, R.D. (1992). Segmentation of Music Primitives. Proceedings of the British Machine Vision Conference, 472–480.
Ng, K.C. (1995). Automated Computer Recognition of Music Scores. Ph.D. Thesis, School of Computing, University of Leeds, UK.
Ng, K.C., Boyle, R.D., & Cooper, D. (1996). Automatic Detection of Tonality Using Note Distribution. Journal of New Music Research, 25(4): 369–381.
Ng, K.C., & Boyle, R.D. (1996). Reconstruction of Music Scores from Primitives Subsegmentation. Image and Vision Computing.
Ng, K.C., Cooper, D., Stefani, E., Boyle, R.D., & Bailey, N. (1999). Embracing the Composer: Optical Recognition of Hand-written Manuscripts. Proceedings of the International Computer Music Conference (ICMC'99) – Embracing Mankind, Tsinghua University, Beijing, China, 500–503.
Ng, K.C., & Cooper D. (2000). Enhancement of Optical Music Recognition using Metric Analysis. Proceedings of the XIII CIM 2000 – Colloquium on Musical Informatics, Italy.
Ng, K.C. (2001). Music Manuscript Tracing. Proceedings of the Fourth IAPR International Workshop on Graphics Recognition (GREC 2001), Canada, 470–481.
Ng, K. C. (2002). Document Imaging for Music Manuscript. Proceedings of the Sixth World Multiconference on Systemics, Cybernetics and Informatics (SCI 2002), Orlando, USA, XVIII, 546–549.
OMeR (n.d.). Optical Music easy Reader, Myriad Software, Retrieved February 8, 2003, from http://www.myriad-online.com/omer.htm
PhotoScore (n.d.). Neuratron, Retrieved February 8, 2003, from http://www.neuratron.com/photoscore.htm Pinto, J., Vieira, P., Ramalho, M., Mengucci, M., Pina, P., & Muge, F. (2000). Ancient Music Recovery for
Digital Libraries. Fourth European Conference on Research and Advanced Technology for Digital Libraries (ECDL 2000), Lisbon.
Porck, H. J., & Teygeler, R. (2000). Preservation Science Survey: An Overview of Recent Developments in Research on the Conservation of Selected Analogue Library and Archival Materials. Council on Library and Information Resources, Washington, D.C., USA.
Pruslin, D.H. (1966). Automated Recognition of Sheet Music. Doctor of Science dissertation, MIT. Roads, C. (1986). The Tsukuba musical robot. Computer Music Journal, 10(2), 39-43. Roth, M. (1994). An Approach to Recognition of Printed Music, Extended Diploma Thesis, Swiss Federal
Institute of Technology, ETH Zürich, CH-8092, Switzerland.
DE4.7.1 — Coding Images of Music
MUSICNETWORK Project 17
Scorscan (n.d.). npc Imaging. Retrieved August 8, 2002, from http://www.npcimaging.com Selfridge-Field, E. (1994). Optical Recognition of Music Notation: A Survey of Current Work. In Hewlett,
W. B. and Selfridge-Field E. (Eds.), Computing in Musicology: An International Directory of Applications, 9, 109–145.
SharpEye (n.d.). visiv, Retrieved August 8, 2002, from http://www.visiv.co.uk SmartScore (n.d.). Musitek, Retrieved February 8, 2002, from http://www.musitek.com/ Stückelberg, M.V., Pellegrini, C. & Hilario, M. (1997). An Architecture for Musical Score Recognition using
High-Level Domain Knowledge. Proceedings of the Fourth International Conference on Document Analysis and Recognition, 2, 813-8.
Stückelberg, M.V. & Doermann, D. (1999). On Musical Score Recognition using Probabilistic Reasoning. Proceedings of the Fifth International Conference on Document Analysis and Recognition, Bangolore, India.
Suen, C.Y., & Wang, P.S.P. (1994). Thinning Methodologies for Pattern Recognition. Series in Machine Perception and Artificial Intelligence, 8, World Scientific.
Vivaldi Scan (n.d.). VivaldiStudio, Retrieved February 8, 2003, from http://www.vivaldistudio.com/Eng/VivaldiScan.asp
Wächter, W., Liers, J., & Becker, E. (1996). Paper Splitting at the German Library in Leipzig. Development from Craftsmanship to Full Mechanisation. Restaurator, 17, 32–42.
11 OMR Bibliography Akiyama, T. and N. Hagita (1990). "Automated entry system for printed documents." Pattern Recognition
23(11): 1141-54. Alphonce, B., B. Pennycook, et al. (1988). Optical music recognition: A progress report. Proceedings of the
Small Computers in the Arts. Andronico, A. and A. Ciampa (1982). On automatic pattern recognition and acquisition of printed music.
Proceedings of the International Computer Music Conference, Venice, Italy. Anquetil, E., B. Couasnon, et al. (1999). A symbol classifier able to reject wrong shapes for document
recognition systems. Graphics Recognition: Recent advances. A. K. Chhabra and D. Dori. Jaipur, India, Springer-Verlag. LNCS 1941: 209-18.
Anstice, J., T. Bell, et al. (1996). The design of a pen-based musical input system. Proceedings of the Sixth Australian Conference on Computer-Human Interaction.
Aoyama, H. and A. Tojo (1982). "Automatic recognition of music score (in Japanese)." Electronic Image Conference Journal 11(5): 427-35.
Aoyama, H. and A. Tojo (1982). "Automatic recognition of printed music (in Japanese)." Institute of Electronics and Communications Engineers of Japan (IECE) TG PREL82-5: 33-40.
Armand, J. P. (1993). Musical score recognition: A hierarchical and recursive approach. Proceedings of the Second IAPR International Conference on Document Analysis and Recognition, Tsukuba Science City, Japan, IEEE Computer Society Press.
Bacon, R. A. and N. P. Carter (1988). "Recognising music automatically." Physics Bulletin 39: 265. Bainbridge, D. (1991). Preliminary experiments in musical score recognition. Department of Computer
Science. The Kings Buildings, Mayfield Road, Edinburgh, GB, University of Edinburgh. Bainbridge, D. (1994). Optical music recognition: Progress report 1, Department of Computer Science,
University of Canterbury. Bainbridge, D. (1994). A complete optical music recognition system: Looking to the future. Bainbridge, D. (1995). Optical music recognition: Progress report 2, Department of Computer Science,
University of Canterbury. Bainbridge, D. (1996). Optical music recognition: A generalised approach. Second New Zealand Computer
Science Graduate Conference. Bainbridge, D. (1997). Extensible optical music recognition. Christchurch, New Zealand, University of
Canterbury. Bainbridge, D. and T. Bell (1996). An extensible optical music recognition system. Proceedings of the 1996
Bainbridge, D. and T. Bell (2001). "The challenge of optical music recognition." Computers and the Humanities 35: 95-121.
Bainbridge, D. and T. Bell (2003). "a music notation construction engine for optical music recognition." Software - Practice and Experience 33(2): 173-200.
Bainbridge, D. and T. C. Bell (1997). Dealing with superimposed objects in optical music recognition. Sixth International Conference on Image Processing and its Applications (Conf. Publ. No.443).
Bainbridge, D., G. Bernbom, et al. (2001). Digital music libraries: Research and development. Proceedings of the Joint Conference on Digital Libraries.
Bainbridge, D. and N. Carter (1997). Automatic recognition of music notation. Handbook of Character Recognition and Document Image Analysis. H. Bunke and P. Wang, World Scientific: 583-603.
Bainbridge, D. and S. Inglis (1998). Musical image compression. Proceedings of the Data Compression Conference, Utah, USA, IEEE Computer Society Press.
Bainbridge, D., C. G. Nevill-Manning, et al. (1999). Towards a digital library of popular music. ACM Conference on Digital Libraries.
Bainbridge, D. and K. Wijaya (1999). Bulk processing of optically scanned music. Seventh International Conference on Image Processing and Its Applications (Conf. Publ. No.465), University of Manchester, UK.
Barton, L. W. G. (2002). The NEUMES project: Digital Transcription of Medieval Chant Manuscriptis. Second International Conference on WEB Delivering of Music, Darmstadt, Germany, IEEE Computer Society Press.
Baumann, S. (1993). Document recognition of printed scores and transformation into MIDI, Deutsches Forschungszentrum für Künstliche Intelligenz GmbH (DFKI): 24.
Baumann, S. (1995). A simplified attributed graph grammar for high-level music recognition. Proceedings of the Third International Conference on Document Analysis and Recognition, Montreal, Canada.
Baumann, S. and A. Dengel (1992). "Transforming printed piano music into MIDI." Proceedings of the IAPR Workshop on Structural and Syntactic Pattern Recognition: 363-72.
Baumann, S. and K. Tombre (1995). Report of the line drawing and music recognition working group. International Association for Pattern Recognition Workshop on Document Analysis Systems. A. L. Spitz and A. Dengel, World Scientific. 14: 1080-3.
Bellini, P., Bruno I., Nesi P., (2001). Optical music sheet segmentation. First Interntional Conference on Web Delivering of Music, Florence, Italy.
Bellini, P., Bruno, I., Nesi, P., ``An Off-Line Optical Music Sheet Recognition", in Visual Perception of Music Notation: On-Line and Off-Line Recognition, edited by Dr. Susan E. George, Idea Group Publisher, USA, 2004.
Beran, T. (1997). Rozpoznavani notoveho zapisu (In Czech). Prague, Czech Republic, Czech Technical University.
Beran, T. (1999). Rozpoznavani notoveho zapisu (In Czech). Prague, Czech Republic, Czech Technical University.
Beran, T. and T. Macek (1999). "Recognition of printed music score." Machine Learning and Data Mining in Pattern Recognition. First International Workshop, MLDM'99. Proceedings. (Lecture Notes in Artificial Intelligence Vol.1715): 174-9.
Blostein, D. and H. S. Baird (1992). A critical survey of music image analysis. Structured Document Image Analysis. H. S. Baird, H. Bunke and K. Yamamoto. Berlin, Springer-Verlag: 405-34.
Blostein, D. and N. P. Carter (1992). Recognition of Music Notation: SSPR '90 Working Group Report. Structured Document Image Analysis. H. S. Baird, H. Bunke and K. Yamamoto. Berlin, Springer Verlag: 573-4.
Blostein, D. and L. Haken (1990). "Template matching for rhythmic analysis of music keyboard input." Proceedings of 10th International Conference on Pattern Recognition: 767-70.
Blostein, D. and L. Haken (1991). "Justification of printed music." Communications of the ACM: 88-91. Blostein, D. and L. Haken (1999). "Using diagram generation software to improve diagram recognition: A
case study of music notation." IEEE Transactions on Pattern Analysis and Machine Intelligence 21(11): 1121-36.
Bruno, I. (2003) "Analisi di immagini di spartiti musicali: metodi e strumenti per il riconoscimento e
DE4.7.1 — Coding Images of Music
MUSICNETWORK Project 19
l'indicizzazione automatica" ("Image music score analysis: methods and tools for the automatic recognition and indexing") Ph.D Thesys. Department of System and Informatics, Faculty of Engineering, University of Florence, Italy
Bulis, A., R. Almog, et al. (1992). "Computerized recognition of hand-written musical notes." Proceedings of the International Computer Music Conference: 110-2.
Capitaine, T., E. M. Mouaddib, et al. (1995). "Automatic recognition of musical scores." ACCV '95. Second Asian Conference on Computer Vision. Proceedings 1: 422-4.
Carter, N. P. (1989). Automatic recognition of printed music in the context of electronic publishing, University of Surrey.
Carter, N. P. (1992). A new edition of Walton's Façade using automatic score recognition. Advances in Structural and Syntactic Pattern Recognition (Proceedings of International Workshop on Structural and Syntactic Pattern Recognition). H. Bunke, World Scientific: 352-62.
Carter, N. P. (1992). "Segmentation and preliminary recognition of madrigals notated in white mensural notation." Machine Vision and Applications 5(3): 223-30.
Carter, N. P. (1993). A generalized approach to automatic recognition of music scores, Department of Music, Stanford University.
Carter, N. P. (1994). "Conversion of the Haydn symphonies into electronic form using automatic score recognition: a pilot study." Proceedings of SPIE 2181: 279-90.
Carter, N. P. (1994). "Music score recognition: Problems and prospects." Computing in Musicology 9: 152-8.
Carter, N. P. and R. A. Bacon (1990). "Automatic recognition of music notation." Proceedings of the International Association for Pattern Recognition Workshop on Syntactic and Structural Pattern Recognition: 482.
Carter, N. P. and R. A. Bacon (1992). Automatic recognition of printed music. Structured Document Image Analysis. H. S. Baird, H. Bunke and K. Yamamoto. Berlin, Springer-Verlag: 456-65.
Carter, N. P., R. A. Bacon, et al. (1988). "The acquisition, representation and reconstruction of printed music by computer: A review." Computers and the Humanities 22(2): 117-36.
Cho, K. J. and K. E. Cho (1996). "Recognition of piano score using skeletal lines and run-length information." Journal of KISS(C) (Computing Practices) 2(4): 461-73.
Choi, J. (1991). Optical recognition of the printed musical score. Electrical Engineering and Computer Science, University of Illinois at Chicago.
Choudhury, G. S., T. DiLauro, et al. (2000). Optical music recognition system within a large-scale digitization project. International Conference on Music Information Retrieval.
Choudhury, G. S., T. DiLauro, et al. (2001). Strike up the score: Deriving searchable and playable digital formats from sheet music. D-Lib Magazine. 7.
Choudhury, G. S., C. Requardt, et al. (2000). "Digital workflow management: the Lester S. Levy digitized collection of sheet music."
Clarke, A. T., B. M. Brown, et al. (1988). Inexpensive optical character recognition of music notation: A new alternative for publishers. Proceedings of the Computers in Music Research Conference, Bailrigg, Lancaster, UK.
Clarke, A. T., B. M. Brown, et al. (1988). "Using a micro to automate data acquisition in music publishing." Microprocessing and Microprogramming 24: 549-54.
Clarke, A. T., B. M. Brown, et al. (1989). "Coping with some really rotten problems in automatic music recognition." Microprocessing & Microprogramming 27(1-5): 547-50.
Clarke, A. T., B. M. Brown, et al. (1990). Problems to be faced by developers of computer based automatic music recognisers. Proceedings of the International Computer Music Conference, Glasgow, Scotland.
Clarke, A. T., B. M. Brown, et al. (1993). "Recognising musical text." Proceedings of the SPIE 2064: 222-33.
Coüasnon, B. (1991). Réseaux de neurones appliqués à la reconnaissance de partitions musicales, Rapport de DEA, Irisa, Université de Rennes I.
Coüasnon, B. (1996). Formalisation grammaticale de la connaissance a priori pour l'analyse de documents : Application aux partitions d'orchestre. Dixiéme Congrés Reconnaissance des Formes et Intelligence
DE4.7.1 — Coding Images of Music
MUSICNETWORK Project 20
Artificielle, Rennes, France. Coüasnon, B. (1996). Segmentation et reconnaissance de documents guidées par la connaissance a priori :
application aux partitions musicales, Université de Rennes. Coüasnon, B. and J. Camillerapp (1994). Using grammars to segment and recognize music scores.
International Association for Pattern Recognition Workshop on Document Analysis Systems, Kaiserslautern, Germany.
Coüasnon, B. and J. Camillerapp (1995). A way to separate knowledge from program in structured document analysis: Application to optical music recognition. Third International Conference on Document Analysis and Recognition, Montréal, Canada.
Coüasnon, B., J. Camillerapp, et al. (1995). Segmentation et reconnaissance de documents guidées par la connaissance a priori: application aux partitions musicales, Association Francaise pour l'Intelligence Artificielle.
Coüasnon, B., P. Brisset, et al. (1995). Using logic programming languages for optical music recognition. International Conference on the Practical Application of Prolog, Paris, France.
Coüasnon, B. and B. Rétif (1995). Using a grammar for a reliable full score recognition system. Proceedings of the International Computer Music Conference, Banff, Canada.
Coüasnon, B. and B. Rétif (1995). "Utilisation d'une grammaire dans la reconnaissance de partitions d'orchestre." Deuxiémes Journeés d'Informatique Musicale: 143-52.
d'Andecy, V. P. (1993). Segmentation et reconnaissance optique de partitions musicales. Rennes, France, IRISA/INSA.
d'Andecy, V. P., J. Camillerapp, et al. (1994). Kalman filtering for segment detection: application to music scores analysis. Proceedings of the 12th IAPR International Conference on Pattern Recognition, Jérusalem, Israel.
d'Andecy, V. P., J. Camillerapp, et al. (1994). Détecteur robuste de segments: Application à l'analyse de partitions musicales. Actes 9 ème Congrés AFCET Reconnaissance des Formes et Intelligence Artificielle, Paris, France.
Di Riso, D. (1992). Lettura automatica di partiture musicali, Università di Salerno, Italy: Università di Salerno, Italy.
Diener, G. R. (1990). "Modeling music notation: A three-dimensional approach." Ph.D: Thesis, Stanford University.
Distasi, R. and e. al. (1993). "Automatic system for reading scores." Proceedings of the 8th Scandinavian Conference on Image Analysis: 1307-10.
Distasi, R., M. Nappi, et al. (1993). "An automatic system for reading musical scores." Proceedings of the 8th Scandinavian Conference on Image Analysis: 1307-10 vol.2.
Droettboom, M., I. Fujinaga, et al. (2002). Optical music interpretation. Proceedings of the Statistical, Structural and Syntactic Pattern Recognition Conference.
Droettboom, M., K. MacMillan, et al. (2002). Using Gamera framework for the recognition of cultural heritage materials. Proceedings of the Joint Conference on Digital Libraries.
Fahmy, H. (1991). A graph-grammar approach to high-level music recognition. Department of Computing and Information Science. Kingston, Ontario, Canada, Queen's University.
Fahmy, H. and D. Blostein (1991). A graph grammar for high-level recognition of music notation. Proceedings of First International Conference on Document Analysis and Recognition, Saint-Malo, France.
Fahmy, H. and D. Blostein (1992). "Graph grammar processing of uncertain data." Proceedings of International Workshop on Structural and Syntactic Pattern Recognition: 373-82.
Fahmy, H. and D. Blostein (1992). Graph grammar processing of uncertain data. Advances in Structural and Syntactic Pattern Recognition. H. Bunke, World Scientific: 373-84.
Fahmy, H. and D. Blostein (1993). "A graph grammar programming style for recognition of music notation." Machine Vision and Applications 6(2-3): 83-99.
Fahmy, H. and D. Blostein (1994). "A graph-rewriting approach to discrete relaxation: Application to music recognition." Proceedings of the SPIE 2181: 291-302.
Fahmy, H. and D. Blostein (1998). "A graph-rewriting paradigm for discrete relaxation: Application to sheet-music recognition." International Journal of Pattern Recognition and Artificial Intelligence 12(6):
DE4.7.1 — Coding Images of Music
MUSICNETWORK Project 21
763-99. Ferrand, M. and A. Cardoso (1998). Scheduling to reduce uncertainty in syntactical music structures.
Advances in Artificial Intelligence. 14th Brazilian Symposium on Artificial Intelligence (SBIA'98). F. M. d. Oliveira, Springer-Verlag. LNAI 1515: 249-58.
Ferrand, M., J. A. Leite, et al. (1999). Improving optical music recognition by means of abductive constraint logic programming. Progress in Artificial Intelligence. 9th Portuguese Conference on Artificial Intelligence, EPIA'99. Proceedings. P. Barahona and J. J. Alferes, Springer-Verlag. LNAI 1695: 342-56.
Ferrand, M., J. A. Leite, et al. (1999). Hypothetical reasoning: An application to optical music recognition. Proceedings. of the APPIA-GULP-PRODE'99 Joint Conference on Declarative Programming, L'Aquila, Italy.
Fischer, K. N. (1978). Computer recognition of engraved music, University of Tennessee. Fletcher, L. A. and R. Kasturi (1988). "A robust algorithm for text string separation from mixed text/graphics
images." IEEE Transactions on Pattern Analysis and Machine Intelligence 10(6): 910-8. Fluhr, C. and J. Abouassly (1989). Music pattern recognition. Proceedings of a workshop held in Toulouse,
September 1988, EEC Concerted Action on "Technology and Blindness", Toulouse. Fotinea, S.-E., G. Giakoupis, et al. (2000). An optical notation recognition system for printed music based on
template matching and high level reasoning. The 6th Recherche d'Informations Assistée par Ordinateur, Paris.
Fujimoto, Y. a. o. (1985). The keyboard playing robot WABOT-2. Bulletin of Science and Engineering Research Laboratory. 112.
Fujinaga, I. (1988). Optical music recognition using projections. Faculty of Music. Montréal, Canada, McGill University: 67 leaves.
Fujinaga, I. (1992). An optical music recognition system which learns. Enabling Technologies for High-Bandwidth Applications, Boston, Massachusetts, USA, SPIE - The International Society for Optical Engineering.
Fujinaga, I. (1996). "Exemplar-based learning in adaptive optical music recognition system." Proceedings of the International Computer Music Conference: 55-6.
Fujinaga, I. (1997). Adaptive optical music recognition. Montreal, Canada, McGill University: viii, 127 leaves.
Fujinaga, I. (2001). Adaptive optical music recognition. 16th Inthernational Congress of the International Musicological Society (1997). Oxford, Oxford University Press.
Fujinaga, I., B. Alphonce, et al. (1992). Optical music recognition on NeXT workstation. Second International Conference on Music Perception and Cognition.
Fujinaga, I., B. Alphonce, et al. (1989). "Issues in the design of an optical music recognition system." Proceedings of the International Computer Music Conference: 113-6.
Fujinaga, I., B. Alphonce, et al. (1992). Interactive optical music recognition. Proceedings of the International Computer Music Conference, San Jose, USA.
Fujinaga, I., B. Alphonce, et al. (1989). "Optical recognition of music notation by computer." Computers in Music Research 1: 161-4.
Fujinaga, I., B. Alphonce, et al. (1991). Optical music recognition: Progress report. Proceedings of the International Computer Music Conference, Montréal, Canada.
Fujinaga, I., S. Moore, et al. (1999). Implementation of exemplar-based learning model for music cognition. Music, mind and science. S. W. Yi. Seoul, Seoul National University Press.
Fujinaga, I., B. Pennycook, et al. (1989). Computer recognition of musical notation. Proceedings of the First International Conference on Music Perception and Cognition, Kyoto, Japan.
Fujinaga, I., B. Pennycook, et al. (1991). "The optical music recognition project." Computers in Music Research 3: 139-42.
Fujinaga, I. and J. Riley (2002). Digital image capture of musical scores. Third International Conference on Music Information Retrieval, Paris, France.
Geggus, K. M. and E. C. Botha (1993). "A model-based approach to sheet music recognition." Elektron 10(1): 25-9.
Gezerlis, V. G. and S. Theodoridis (2002). "Optical character recognition of the Orthodox Hellenic
DE4.7.1 — Coding Images of Music
MUSICNETWORK Project 22
Byzantine Music notation." Pattern Recognition 35: 895-914. Glass, S. (1989). Optical music recognition. Department of Computer Science. Christchurch, New Zealand,
University of Canterbury. Goolsby, T. W. (1994). "Eye movement in music reading: Effects of reading ability, notational complexity,
and encounters." Music Perception 12(1): 77-96. Goolsby, T. W. (1994). "Profiles of processing: Eye movements during sightreading." Music Perception
12(1): 97-123. Hachimura, K. and Y. Ohno (1987). "A system for the representation of human body movements from
dance scores." Pattern Recognition Letters 5: 1-9. Hewlett, W. B. and E. Selfridge-Field (1990). "Optical recognition of musical data." Computing in
Musicology: A Directory of Research: 36-45. Homenda, W. (1995). "Optical pattern recognition for printed music notation." Proc. SPIE - Int. Soc. Opt.
Eng. (USA), Proceedings of the SPIE - The International Society for Optical Engineering 2490: 230-9.
Homenda, W. (1996). "Automatic recognition of printed music and its conversion into playable music data." Control and Cybernetics 25(2): 353-67.
Hori, T., S. Wada, et al. (1999). "Automatic music score recognition/play system based on decision based neural network." 1999 IEEE Third Workshop on Multimedia Signal Processing (Cat. No.99TH8451): 183-4.
Inokuchi, S. (1981). "Musical database." Journal of the Institute of Electronics and Communication Engineers of Japan 64(5): 466-8.
Inokuchi, S. and H. Katayose (1990). "Computer and music." Journal of the Institute of Electronics, Information and Communication Engineers 73(9): 965-7.
Itagaki, T. S., S. Hashimoto, et al. (1990). "Automatic recognition on some different types of musical notation." Proceedings of the International Association for Pattern Recognition Workshop on Syntactic and Structural Pattern Recognition: 488 ff.
Itagaki, T. S., S. Hashimoto, et al. (1992). Automatic recognition of several types of musical notation. Structured Document Image Analysis. H. S. Baird, H. Bunke and K. Yamamoto. Berlin, Springer-Verlag: 466-76.
Kassler, M. (1970). An essay toward specification of a music-reading machine. Musicology and the computer. B. S. Brook, City University of New York Press: B.
Kassler, M. (1972). "Optical character recognition of printed music: A review of two dissertations." Perspectives of New Music 11(2): 250-4.
Katayose, H., T. Fukuoka, et al. (1990). Expression extraction in virtuoso music performances. Proceedings of the Tenth International Conference on Pattern Recognition, Atlantic City, New Jersey.
Katayose, H. and S. Inokuchi (1989). "The kansei music system." Computer Music Journal 13(4): 72-7. Katayose, H., H. Kato, et al. (1989). "An approach to an artificial music expert." Proceedings of the
International Computer Music Conference: 139-46. Kato, H. and S. Inokuchi (1988). "Automatic recognition of printed piano music based on bar unit processing
(in Japanese)." Transactions of I. E. C. E. J71-D(5): 894-901. Kato, H. and S. Inokuchi (1990). The recognition system for printed piano music using musical knowledge
and constraints. Proceedings of the IAPR Workshop on Syntactic and Structural Pattern Recognition, Murray Hill, New Jersey.
Kato, H. and S. Inokuchi (1992). A recognition system for printed piano music using musical knowledge and constraints. Structured Document Image Analysis. H. S. Baird, H. Bunke and K. Yamamoto. Berlin, Springer-Verlag: 435-55.
Kim, W. J., M. J. Chung, et al. (1987). "Recognition system for a printed music score." Proceedings of TENCON 87: 1987 IEEE Region 10 Conference 'Computers and Communications Technology Toward 2000 2: 573-7.
Kinoshita, T., H. Muraoka, et al. (1998). "Note recognition using statistical information of musical note transitions." Journal of the Acoustical Society of Japan 54(3): 190-8.
Kobayakawa, T. (1993). Auto music score recognition system. Proceedings SPIE: Character Recognition Technologies.
DE4.7.1 — Coding Images of Music
MUSICNETWORK Project 23
kopec, G., P. chou, et al. (1996). "Markov source model for printed music decoding." Journal of Electronic Imaging 5(1): 7-14.
Kopec, G. E., P. A. Chou, et al. (1995). "Markov source model for printed music decoding." Proc. SPIE - Int. Soc. Opt. Eng. (USA), Proceedings of the SPIE - The International Society for Optical Engineering 2422: 115-25.
Lee, M. W. and J. S. Choi (1985). "The recognition of printed music score and performance using computer vision system (in Korean and English translation)." Journal of the Korean Institute of Electronic Engineers 22(5): 429-35.
Lee, S. and J. Shin (1994). "Recognition of music scores using neural networks." Journal of the Korea Information Science Society 21(7): 1358-66.
Lee, S. D. (1995). Automatic Optical Music Recognition. Hong Kong, Hong Kong University, Department of Computer Science: 28.
Leite, J. A. and M. Ferrand (1994). RIEM: Reconhecimento e Interpretação de Escrita Musical (in Portuguese). Dept. de Engenharia Electrotécnica, Faculdade de Ciências e Tecnologia, Universidade de Coimbra.
Leite, J. A., M. Ferrand, et al. (1998). RIEM: A system for recognition and interpretation of music writing (in Portuguese), Dept. Engenharia Informatica, Faculdade de Ciências e Tecnologia, Universidade de Coimbra.
Leplumey, I. and J. Camillerapp (1991). "Comparison of region labelling for musical scores." Proceedings of First International Conference on Document Analysis 2: 674-82.
Leplumey, I. and J. Camillerapp (1991). "Coopération entre la segmentation des régions blanches et des régions noires pour l'analyse de partitions musicales." AFCET, 8e Congress Reconnaissance des Formes et Intelligence Artificielle 3: 1045-52.
Leplumey, I., J. Camillerapp, et al. (1993). A robust detector for music staves. Proceedings of the International Conference on Document Analysis and Recognition, Tsukuba Science City, Japan.
Luth, N. (2002). Automatic identification of music notations. Second International Conference on WEB Delivering of Music, Darmsradt, Germany, IEEE Computer Society Press.
MacMillan, K., M. Droettboom, et al. (2001). Gamera: A structured document recognition application development environment. Proceedings of the International Symposium on Music Information Retrieval.
MacMillan, K., M. Droettboom, et al. (2002). Gamera: Optical music recognition in a new shell. International Computer Music Conference, Sweden.
Maenaka, K. and Y. Tadokoro (1983). "Recognition of music using the special image-input-device enabling to scan the staff of music as the supporting system for the blind (in Japanese)." Prl83-60: 37-45.
Mahoney, J. V. (1982). Automatic analysis of musical score images, Massachusetts Institute of Technology. Marinai, S. and P. Nesi (1999). Projection based segmentation of musical sheets. 5th International
Conference on Document Analysis and Recognition, Bangalore, India, IEEE press. Martin, N. G. (1987). Towards computer recognition of the printed musical score, Thames Polytechnic. Martin, P. (1989). Reconnaissance de partitions musicales et réseaux de neurones: une étude. Actes 7 iéme
Congrés AFCET de Reconnaissance des Formes et Intelligence Artificielle. Martin, P. (1992). Réseaux de neurones artificiels : Application à la reconnaissance optique de partitions
musicales, IMAG, Grenoble, France. Martin, P. and C. Bellissant (1991). Neural networks at different levels of musical score image analysis
system. Proceedings of 7th Scandinavian Conference on Image Analysis, Aalborg, Denmark. Martin, P. and C. Bellissant (1991). "Low-level analysis of music drawing images." Proceedings of the
International Conference on Document Analysis and Recognition: 417-25. Martin, P. and C. Bellissant (1992). "Neural networks for the recognition of engraved musical scores."
International Journal of Pattern Recognition and Artificial Intelligence 6(1): 193-208. Matsushima, T. (1988). "Automatic printed-music-to-braille translation system." Journal of Information
Processing 11(4): 249-57. Matsushima, T. (1992). "Computerized Japanese traditional music processing system." Proceedings of the
International Computer Music Conference: 121-4. Matsushima, T., T. Harada, et al. (1985). Automated recognition system for musical score: The vision
DE4.7.1 — Coding Images of Music
MUSICNETWORK Project 24
system of WABOT-2. Bulletin of Science and Engineering Research Laboratory, Waseda University.
Matsushima, T., S. Ohteru, et al. (1989). "An integrated music information processing system." Proceedings of the International Computer Music Conference: 191-8.
Matsushima, T., S. Ohteru, et al. (1985). "Automatic recognition of printed music (in Japanese)." Japan Acoustics Society Journal 41(6): 412-5.
Matsushima, T., I. Sonomoto, et al. (1985). Automated high speed recognition of printed music (WABOT-2 vision system). Proceedings of the International Conference on Advanced Robotics (ICAR), Shiba Koen Minato-ku, Tokyo.
McGee, W. and P. Merkley (1991). "The optical scanning of medieval music." Computers and the Humanities 25(1): 47-53.
McGee, W. F. (1994). "MusicReader: An interactive optical music recognition system." Computing in Musicology 9: 146-51.
McGee, W. F. and P. Merkley (1989). "Optical recognition of music using page straightening." McLean, G. I. (1991). Music recognition. Department of Computer Science, Heriot-Watt University. McPherson, J. R. (2002). Introducing feedback into an optical music recognition system. Third International
Conference on Music Information Retrieval, Paris, France. McPherson, J. R. and D. Bainbridge Coordinating knowledge within an optical music recognition system. Miyao, H. (1997). Printed score recognition using neural networks, Shinshu University. Miyao, H. (2002). Stave extraction for printed music scores. Third International Conference on Intelligent
Data Engineering and Automated Learning (IDEAL2002). Miyao, H. and R. M. Haralick (2000). Format of ground truth data used in the evaluation of the results of an
optical music recognition system. IAPR Workshop on Document Analysis Systems. Miyao, H. and Y. Nakano (1995). "Head and stem extraction from printed music scores using a neural
network approach." Proceedings of the Third International Conference on Document Analysis and Recognition 2: 1074-9.
Miyao, H. and Y. Nakano (1996). "Note symbol extraction for printed piano scores using neural networks." IEICE Transactions on Information and Systems E79-D(5): 548-54.
Miyao, H. T., T. Ejima, et al. (1990). "Recognition for printed piano scores (in Japanese)." Nlc90-34, Pru90-74: 39-46.
Miyao, H. T., T. Ejima, et al. (1992). "Symbol recognition for printed piano scores based on the musical knowledge (in Japanese)." Transactions of the Institute of Electronics, Information and Communication Engineers D-II J75D-II(11): 1848-55.
Modayur, B. R. (1991). Restricted domain music score recognition using mathematical morphology. Fifth International Conference on Symbolic and Logical Computing.
Modayur, B. R. (1996). Music score recognition - a selective attention approach using mathematical morphology. Seattle, University of Washington, Electrical Engineering Department.
Modayur, B. R., R. M. Haralick, et al. (1992). "On printed music score symbol recognition." Proceedings of Symposium on Document Analysis and Information Retrieval: 16-8.
Modayur, B. R., V. Ramesh, et al. (1992). MUSER-a prototype musical recognition system using mathematical morphology. Seattle, Intelligent Systems Laboratory, EE Dept, FT-10, University of Washington: 32.
Modayur, B. R., V. Ramesh, et al. (1993). "MUSER: a prototype musical recognition system using mathematical morphology." Machine Vision and Applications 6(2-3): 140-50.
Musitek (1994). "Musitek, Midiscan." Keyboard 20(3): 136. Nagy, G. (1989). "Document analysis and optical character recognition." Proceedings of the 5th International
Conference on Image Analysis and Processing: 511-29. Nakamura, Y., M. Shindo, et al. (1978). "Input method of [musical] note and realization of folk music data-
base (in Japanese)." Institute of Electronics and Communications Engineers of Japan (IECE) TG PRL78-73: 41-50.
Nelson, G. and T. R. Penney (1973). Pattern recognition in musical score - Project no. M88. Computers and the Humanities. 8: 50-1.
DE4.7.1 — Coding Images of Music
MUSICNETWORK Project 25
Newell, C. a. and W. Homenda (1993). MIDISCAN for windows. Ng, K. C. (1995). Automated computer recognition of music score. School of Computer Studies. Leeds, GB,
University of Leeds. Ng, K. C. (2001). Music manuscript tracing. Fourth IAPR International Workshop on Graphics Recognition,
Ontario, Canada. Ng, K. C. (2002). Music manuscript tracing. Graphics Recognition: Algorithms and Applications, Springer-
Verlag. 2390: 330-342. Ng, K. C. (2002). Document imaging for music manuscript. 6th World Multiconference on Systemics,
Cybernetics and Informatics, Florida, USA. Ng, K. C. (2002). Optical music analysis: A reverse engineering approach. EVA 2002 Florence, Italy. Ng, K. C. and R. D. Boyle (1992). "Segmentation of music primitives." BMVC92. Proceedings of the British
Machine Vision Conference: 472-80. Ng, K. C. and R. D. Boyle (1994). Reconstruction of music scores from primitive sub-segmentation, School
of Computer Studies, University of Leeds. Ng, K. C. and R. D. Boyle (1996). "Recognition and reconstruction of primitives in music scores." Image
and Vision Computing 14(1): 39-46. Ng, K. C., R. D. Boyle, et al. (1995). "Low- and high-level approaches to optical music score recognition."
IEE Colloquium on Document Image Processing and Multimedia Environments (Digest No.1995/191) 3: 1-6.
Ng, K. C., R. D. Boyle, et al. (1995). Automated optical musical score recognition and its enhancement using high-level musical knowledge. XI Colloquium on Musical Informatics.
Ng, K. C., R. D. Boyle, et al. (1996). Hand written music manuscript recognition. Proceedings of the International Computer Music Conference.
Ng, K. C. and D. Cooper (2000). Enhancement of optical music recognition using metric analysis. XIII Colloquium on Musical Informatics.
Ng, K. C., D. Cooper, et al. (2001). Towards an integrated handwritten music, manuscript analysis and recognition system. New Synergies in Digital Creativity Conference for Content Integrated Research in Creative User Systems, Glasgow, UK.
Ng, K. C., D. Cooper, et al. (1999). "Embracing the composer: Optical recognition of hand-written manuscripts." Proceedings of the International Computer Music Conference: 500-3.
Ohteru, S. (1987). "Automatic recognition of music score (in Japanese)." Bit (special issue on Computer and Music): 92-100.
Ohteru, S. (1988). "Data entry and automatic recognition of music score (in Japanese)." Journal of the Information Processing Society of Japan 29(6): 586-92.
Ohteru, S. and et al. (1984). "A multi processor system for high speed recognition of printed music (in Japanese)." National Convention Records of I. E. C. E.
Ohteru, S., T. Matsushima, et al. (1985). "Automatic recognition of printed music (in Japanese)." Japan Acoustics Society Journal 41(6): 412-5.
Onoe, M., M. Ishizuka, et al. (1979). "Experiment on automatic music reading (in Japanese)." Proceedings of 20th IPSJ National Conference: 6F-5.
Ostenstad, B. (1988). "Oppdeling av abjektene i et digitalt notebilde i klassifiserbare enheter (in Norwegian)." Institute of Informatics.
Pennycook, B. (1990). "Towards advanced optical music recognition." Advanced Imaging 27: 54-7. Perrotti, F. A. and R. A. Lotufo (1993). "Pre-processamento, Exctracao de Atributos e Primeiro Nivel de
Classiccao para un Sistema de Reconhecimento Otico de Simbolos Musicais." [Preprocessing, Feature Extraction, and First Classification Level for an Optical Recognition System.] in VI Brazilian Symposium in Computer Graphics and Image Processing, SINGRAPI.
Pinto, J., P. Vieira, et al. (2000). Ancient music recovery for digital libraries. Fourth European Conference on Research and Advanced Technology for Digital Libraries (ECDL 2000), Lisbon.
Pinto, J. C., P. Vieira, et al. (2003). "A new graph-like classification method applied to ancient handwritten musical symbols." International Journal of Document Analysis and Recognition 6(1): 10-22.
Prerau, D. S. (1970). Computer pattern recognition of standard engraved music notation, Massachusetts Institute of Technology.
DE4.7.1 — Coding Images of Music
MUSICNETWORK Project 26
Prerau, D. S. (1971). "Computer pattern recognition of printed music." AFIP Conference proceedings of the 1971 fall joint computer conference 39: 153-62.
Prerau, D. S. (1975). "Do-Re-Mi: A program that recognizes music notation." Computers and the Humanities 9(1): 25-9.
Pruslin, D. (1966). Automatic recognition of sheet music, Massachusetts Institute of Technology. Randriamahefa, R. J., J. P. Cocquerez, et al. (1993). Printed music recognition. Proceedings of the Second
IAPR International Conference on Document Analysis and Recognition, Tsukuba Science City, Japan.
Reed, K. T. (1995). Optical music recognition, University of Calgary, Canada. Reed, K. T. and J. R. Parker (1996). "Automatic computer recognition of printed music." Proceedings of the
13th International Conference on Pattern Recognition 3: 803-7. Richard, D. M. (1990). "Godel tune: formal models in music recognition systems." ICMC Glasgow 1990.
Proceedings: 338-40. Roach, J. W. and J. E. Tatem (1988). "Using domain knowledge in low-level visual processing to interpret
handwritten music: an experiment." Pattern Recognition 21(1): 33-44. Roads, C. (1986). "The Tsukuba musical robot." Computer Music Journal 10(2): 39-43. Rossant, F. (2002). "A global method for music symbol recognition in typeset music sheets." Pattern
Recognition Letters 23: 1129-41. Rossant, F. and I. Bloch (2001). Reconnaissance de partitions msicales par modélisation floue et intégration
de règles musicales. Proceedings of GRETSI 2001, Toulouse, France. Roth, M. (1992). OMR-optical music recognition. Institute for theoretical computer science ETH Zurich,
CH-8092. Zurich, Switzerland, Swiss Federal Institute of Technology. Roth, M. (1994). An approach to recognition of printed music. Department of Computer Science. Zurich,
Switzerland, Swiss Federal Institute of Technology. Ruttenberg, A. (1991). Optical reading of typeset music. Department of Architecture and Planning,
Massachusetts Institute of Technology: 162 p. Sawada, H., T. Matsushima, et al. (1990). "A practical bilateral translation system between printed music and
braille." Proceedings of 6th International Workshop on Computer Applications for the Visually Handicapped.
Sawaki, M., H. Murasei, et al. (1998). "A study on SYAKUHACHI score recognition with embedded symbols." Transactions of the Institute of Electronics, Information and Communication Engineers D-II J81D-II(10): 2480-2.
Seales, W. B. and A. Rajasekar (1995). Interpreting music manuscripts: a logic-based, object-oriented approach. Image Analysis Applications and Computer Graphics. Third International Computer Science Conference. ICSC95. Proceedings, Hong Kong.
Selfridge-Field , E. (1994). "Optical recognition of musical notation: A survey of current work." Computing in Musicology 9: 109-45.
Sicard, E. (1992). "An efficient method for the recognition of printed music." Proceedings of 11th International Conference on Pattern Recognition (IAPR): 573-6.
Sokei, S., T. Yamashiro, et al. (1997). Study of recognition for okinawa syamisen score KUNKUNSI. Bulletin of the Faculty of Engineering, University of the Ryukyus. 53: 77-82.
Sonomoto, I., T. Harada, et al. (1985). "Automated recognition system of printed music for playing keyboards (in Japanese)." Acoustical Society of Japan: TG MA84-22 17-22.
Stevens, C. and C. Latimer (1992). "A comparison of connectionist models of music recognition and human performance." Minds and Machines 2(4): 379-400.
Stückelberg, M. V. (1999). Musical score recognition using probabilistic inference, University of Maryland. Stückelberg, M. V. and D. Doermann (1999). On musical score recognition using probabilistic reasoning.
Proceedings of the Fifth International Conference on Document Analysis and Recognition., Bangolore, India.
Stückelberg, M. V., C. Pellegrini, et al. (1997). "An architecture for musical score recognition using high-level domain knowledge." Proceedings of the Fourth International Conference on Document Analysis and Recognition (Cat. No.97TB100138) 2: 813-8.
Stückelberg, M. V., C. Pellegrini, et al. (1997). A preview of an architecture for musical score recognition.
DE4.7.1 — Coding Images of Music
MUSICNETWORK Project 27
Sumsion, H. (1992). Optical music recognition. Leeds, University of Leeds: 91. Thorud, E. (1988). Analyse av notebilder (in Norwegian). Institute of Informatics P.O. Box 1080 Blindern,
N-0316 Oslo 3, Norway. Tojo, A. and H. Aoyama (1982). Automatic recognition of music score. Proceedings of 6th International
Conference on Pattern Recognition, Germany. Tonnesland, S. (1986). "SYMFONI: System for note coding (in Norwegian)." Institute of Informatics. Vieira, P., A. Marcolino, et al. (2000). Ancient music optical recognition. 11th Portuguese Conference on
Pattern Recognition RecPad2000. Vieira, P. and J. Pinto (2001). Recognition of musical symbols in ancient manuscripts. International
Conference on Image Processing. Vieira, P. and J. Pinto (2001). A classification method for ancient hand-written musical symbols. Actas de
10° Encontro Português de Computação Gráfica, Lisbon. Vieira, P., M. Ramalho, et al. (2000). Ancient music optical recognition. V Ibero-American Symposium on
Pattern Recognition, Lisbon. Watkins, G. (1994). "A fuzzy syntactic approach to recognising hand-written music." Proceedings of the
International Computer Music Conference: 297-302. Wijaya, K. and D. Bainbridge (1999). "Staff line restoration." Seventh International Conference on Image
Processing and Its Applications (Conf. Publ. No.465) 2: 760-4. Wilk, R. (1995). Converting graphic musical data to a machine playable form. Music. Montreal, McGill
University. Wilkinson, J. (1995). Optical music recognition. School of Computer Science. Birmingham, University of
Birmingham: 44. Wittlich, G. (1973). "Project SCORE." Computational Musicology Newsletter 1(1): 6. Wittlich, G. (1973). Project SCORE. International Conference on Computers in the Humanities. Wittlich, G., D. Byrd, et al. (1978). "A system for interactive encoding of music scores under computer
control." Computers and the Humanities 12(4): 309-19. Wittlich, G. E. (1974). Non-physics measurements on the PEPR System: Seismograms and music scores.
Report to the Oxford Conference on Computer Scanning, Oxford Nuclear Physics Laboratory: 487-9.
Wolman, A., J. Choi, et al. (1992). "Recognition of handwritten music notation." Proceedings of the International Computer Music Conference: 125-7.
Wolman, A. and T. Yeager (1994). Optical music recognition: progress report. Proceedings of the International Computer Music Conference.
Wright, D. J. (1994). Optical music recognition: A deterministic, object-oriented approach. Department of Computer Science. Victoria, BC, University of Victoria: 233.
Yadid, O., E. Brutman, et al. (1992). "RAMIT: Neural network for recognition of musical notes." Proceedings of the International Computer Music Conference: 128-31.
Yadid-Pecht, O., M. Gerner, et al. (1996). "Recognition of handwritten musical notes by a modified neocognitron." Machine Vision and Applications 9(2): 65-72.
Yin, F., G. Qingshi, et al. (1989). "Principle on designing the music reading system (in Chinese)." Mini-Micro Systems 10(12): 1-10.
Yoda, I., K. Yamamoto, et al. (1995). Automatic construction of recognition procedures for musical notes by genetic algorithm. International Association for Pattern Recognition Workshop on Document Analysis Systems. A. L. Spitz and A. Dengel, World Scientific. 14: 225-36.
